Volume 43, Issue 4

This study examined production and perception of the contrast between /t/ –/d/ by subjects differing in native language and age. Acoustic analysis revealed that native speakers of English realized word-initial /t/ with significantly longer VOT values (approximately 85 ms) than native speakers of Spanish (approximately 19 ms). Native English and Spanish adults realized /t/ with VOT values that were non-significantly longer than those of 9-year-olds of the same native language background. Native English adults prevoiced /d/ more often than English children, but Spanish adults and children realized /d/ with lead VOT values in nearly every instance. In a labelling task, the native English speakers showed steeper identification functions, and category boundaries at significantly longer VOT values, than age-matched native Spanish subjects. The boundaries of both the native English and Spanish adults occurred at significantly longer VOT values than those of children who spoke the same language. Three possible explanations are offered for the effect of age on stop voicing judgments: the auditory processing of acoustic parameters associated with stops changes with age; listeners require an increasingly long VOT interval to perceive /t/ as they get older because they produce /t/ with somewhat longer VOT values; listeners ‘attune’ their perception to optimally match the stops they hear.

Changes in stress and rate of speech have been shown to have different effects on the durations of speech intervals. We examined the effects of such changes on syllable and intrasyllabic segment durations in the word-initial syllable of three-syllable words in Modern Greek, and in conflict with previously reported results for other languages, found rate and stress to have equal effects. We propose that languages may differ in terms of temporal programming in the same sense as they differ in terms of syntactic or phonological rule systems.

Triadically collected judgments on the orosensory similarity of Dutch vowels were studied in terms of three phonological and phonetic descriptions of the Dutch vowel system, differing mainly in their respective treatments of vowel heights. A feature matrix using four vowel heights rather than three had the greatest resemblance with a three-dimensional representation of the vowels on the basis of orosensory judgments. A hierarchical representation of the data indicates that the feature front/back outweighs the features high/low and long/short as far as their relative impact is concerned. Roundedness appeared to play no dominant role in the judgment process.

This paper presents a challenge to the well-established monophonematic view of German and English diphthongs. Empirical evidence is provided by speech error patterns showing that misordered diphthongs occasionally split apart during the error process. Since both splitting up and remaining intact can be observed with diphthongs, a linear analysis, which predicts an all-or-none behaviour, proves to be inadequate. An explanation for the relative cohesiveness of these phonemes can be found in a multilevel representation within which diphthongs are two discrete atoms on the segmental level which are molecularized in the suprasegmental structure.